AccD: A Compiler-based Framework for Accelerating Distance-related Algorithms on CPU-FPGA Platforms

TL;DR

AccD is a compiler-based framework that unifies and accelerates distance-related algorithms on CPU-FPGA platforms, achieving significant speedups and energy efficiency improvements.

Contribution

It introduces a domain-specific language and optimizing compiler to seamlessly integrate algorithmic and hardware optimizations for distance-related algorithms.

Findings

Achieves 31.42x speedup over CPU implementations

Attains 99.63x better energy efficiency

Eases deployment on mainstream CPU-FPGA platforms

Abstract

As a promising solution to boost the performance of distance-related algorithms (e.g., K-means and KNN), FPGA-based acceleration attracts lots of attention, but also comes with numerous challenges. In this work, we propose AccD, a compiler-based framework for accelerating distance-related algorithms on CPU-FPGA platforms. Specifically, AccD provides a Domain-specific Language to unify distance-related algorithms effectively, and an optimizing compiler to reconcile the benefits from both the algorithmic optimization on the CPU and the hardware acceleration on the FPGA. The output of AccD is a high-performance and power-efficient design that can be easily synthesized and deployed on mainstream CPU-FPGA platforms. Intensive experiments show that AccD designs achieve 31.42x speedup and 99.63x better energy efficiency on average over standard CPU-based implementations.